Introduction
In this article, based on a presentation given by the author at the First INDICARE Workshop on Business Models for Mobile Music and DRM, 30 September, 2004, Berlin, we examine the features and advantages of P2P networks with respect to major constituencies in digital content value chains: consumers, the law, content owners, and technology developers. We then show how early, mostly US-based legitimate online content services have grown to appropriate some P2P features (and vice versa – how some P2P-derived services are emerging that purport to respect copyright).

We go on to analyze the likelihood of various P2P features making it into copyright-respecting services, and we assess features of P2P that are likely to remain largely absent from legitimate services, and by suggesting trends that will persist into the future.

The good and bad of P2P
Consumers are attracted to P2P file-sharing services for a host of reasons, in addition to the obvious one (from consumers’ perspective) of not charging for content. P2P has several advantages, including these:

  • Anyone can participate: P2P networks do not respect boundaries, national or otherwise.
  • Render on many devices: P2P networks provide content files that can be rendered on a wide variety of user devices, e.g., MP3 files for music.
  • Permanent files: files available on P2P networks do not “expire”; they are playable indefinitely.
  • Share with friends: there are no restrictions on sending copies of files from P2P networks to friends or acquaintances.
  • Tastemakers: many P2P networks enable users to act as recommenders or tastemakers who can acquire followings among users.
  • Otherwise unavailable content: P2P networks are natural havens for content that is unavailable elsewhere, such as digital “rips” of tracks from out-of-print or obscure music albums.
  • Optimized delivery: some P2P networks, such as BitTorrent, exploit the power of machines attached to the network to divide up the task of sending large files around.
  • Free content: P2P networks can make content available at no charge.
  • Superdistribution: P2P networks can conceivably support Superdistribution, as described below.

At the same time, P2P networks have certain disadvantages, aside from the fact that their use lays consumers open to infringement liability. They are plagued with spoof files, which record companies and other content owners put there in order to degrade the overall service quality. Other files may be incomplete or have poor sound quality. Some file-sharing services make their money by forcing users to view ads or by installing intrusive “spyware” onto their machines. And file-sharing services generally have very limited information about artists and content.

Copyright-respecting services tend not to have any of these problems: they offer guaranteed, complete content with audio/video quality that ranges from decent to excellent, few or no ads, and no spyware. And many copyright-respecting services offer a wealth of artist and content information, recommendations, links, and so on.

Surely there ought to be a way to incorporate some of the desirable features of P2P while still ensuring that copyright owners are compensated – either by adding P2P-like features to copyright-respecting architectures or by adding copy controls onto P2P network architectures.

One general approach to bridging the gap between P2P and existing paid services is known as Superdistribution. Although this term was popularized after the rise of the Internet (Cox 1996), it dates back further (Mori and Kawahara 1990). In this context, it means multi-tiered distribution that starts with the owner of the content and enables entities at each step to redistribute content under their own business terms. Some of the earliest DRM technologies, such as IBM’s infoMarket, attempted to implement multi-tiered distribution with e-commerce, but it was found to be too complex, especially in the days before e-commerce components (e.g., online payment processing) were commonly available.

Yet as we will see, Superdistribution is beginning to experience a comeback as the ramifications of the model for certain types of content are explored. Among other things, Superdistribution can provide a framework that enables tastemakers (see above) to get paid. With general-purpose e-commerce software easily available, it is conceivable to layer Superdistribution on top of P2P network architectures.

Adding P2P features to legitimate services
We can speculate on the likelihood of various features of P2P being added to copyright-respecting networks by looking at how attractive they are to various constituencies:

  • Consumers: is the feature desirable or uninteresting?
  • The law: is it legal or illegal?
  • Content owners: does it make sense from a business perspective or not?
  • Technology: is it easy or difficult to implement with DRM and related technologies?

Table 1 summarizes many of the salient features of P2P networks with respect to the above four constituencies. The salient features are explained below.

 

Desirable for Users

Legal

Acceptable to IP
Owners

Easy with DRM

Likelihood in legitimate Services

Anyone can participate

 

National boundaries

In theory...

 

Not worth the trouble

Render on many devices

 

EU private copying laws

Within limits: products vs.

content

Hardest technology problem

Legal or via 3rd party solutions

Permanent files

 

 

Depends on business model

 

Some business models

Share with friends

 

Generally restricted

Within limits

 

Within limits

Tastemakers

 

 

 

 

 

Otherwise unavailable content

 

Licensing obstacles

If they can get paid

 

Unlikely

Optimized delivery

Marginal importance

 

Marginal importance

Complex but feasible

Through CDNs

Free content

 

No!

No!

 

No!

Superdistribution

Remains to be seen

Licensing contracts

Only in certain cases

Getting easier

Remains to be seen


Table 1: Salient features of P2P networks with respect to four constituencies.
Legend: Green means attractive, yellow denotes reservations or limitations, orange denotes serious reservations/limitations, and red means unattractive or impossible. The rightmost column represents an assessment of how likely the feature in each row is to make it into copyright-respecting content services.

Let’s examine some of the most noteworthy issues implied in the above table.

Anyone Can Participate
The obstacles to anyone participating in a copyright-respecting P2P network are national boundaries that govern e-commerce as well as content licensing agreements. This type of problem is likely to be judged too complex to be worth solving; services will need to remain specific to countries. This is the case today with services that are available in multiple countries, such as Tiscali Music Club, Vodafone live!, iTunes, and Napster.

Play on Many Devices
Technology is the biggest hurdle to a copyright-respecting service providing content that plays on many different devices. Interoperability among formats and DRM schemes is elusive. At this time, attempts at interoperability are coming from various different sources, including putative de facto standards (Microsoft Windows Media), open DRM standards (Open Mobile Alliance Download and DRM, see OMA 2002), open interoperability standards (Digital Media Project, see DMP 2004, Coral Consortium, see Coral 2004), and ad-hoc interoperability (RealNetworks’ Harmony, which is part of its RealPlayer Music Store service). Even more basic problems like interoperability of consumers’ online identities have not been solved yet.

Apart from technology problems, there is a real question of whether content owners are even interested in making their content available on any device. For centuries, content owners have been in the business of selling products, and there is a general mentality among them that users must buy a new product each time they want to consume content on a different device.

Permanent Files
Although consumers are slowly starting to understand the value of subscription services (at the right price point, of course), consumers are still very much behind the idea of “owning” content. Content owners will need to provide permanent downloads for the foreseeable future; many will do so.

Share with Friends
This one is rather ironic. For the most part, the law says that sharing content with “friends” without compensating rights holders is infringement. Private copying laws in some EU countries allow consumers to make copies for the use of themselves or family members, while fair dealing law in the UK (UK Copyright, Design and Patents Act, s. 29, 30 (1988)) empowers courts to render decisions on such matters according to factors like the type of usage and its effect on the overall market for the content. The fair use laws in the US are similar (17 United States Code § 107 (2000)).

Yet laws may well end up not being the limiting factor in this case – because most online sales of content are not really “sales” at all, but rather are license contracts, and thus are not necessarily subject to fair dealing or private copying law restrictions. Moreover, consumers have come to expect some freedom to make copies of content (usually in analogue) for friends and family; therefore, as we will shortly see, expectations are driving market forces so that more and more legitimate online content services support some carefully circumscribed notion of “sharing”.

Tastemakers
While some people in the P2P community are under the impression that this idea was invented there, legitimate content services have been making user recommendations available for quite some time.

Perhaps the first successful “tastemaker” implementation in the media industry was the affiliate network feature of Amazon.com, which enables “affiliates” to create websites (or email messages) with specially coded links to products on Amazon. If a user clicks on such a link on an affiliate website and buys the product, the affiliate earns a sales commission. More recently, Amazon implemented a variation on this theme called Listmania, in which users can create themed lists of recommended products that appear on the site as users browse related items. Earlier this year, iTunes created its own affiliate network through an affiliate network provider called LinkShare.

P2P tastemaker functions do go beyond the above capabilities by providing built-in ways for users to search and browse other users’ collections or recommendations. Yet the larger point is that it is eminently possible for copyright-respecting online content services to offer “tastemaker” features.

Otherwise Unavailable Content
One of the truly great things about P2P file-sharing services is that they give collectors of the rare and obscure chances to show off their collections, so that the material can become less rare and obscure through exposure. Unfortunately, however, many of those rarities are likely to be still under copyright, in which case such aficionado altruism is likely to run afoul of the law. Unfortunately, it is impossible in the general case to solve the nightmarish licensing problems that would come up in this case; such problems are very difficult to solve even in the analogue world.

It is possible that a government might pass a compulsory licensing law that requires content to be made available online under reasonable and non-discriminatory terms, or at least provides fallback terms for content that is not licensable through conventional methods. This would help in many cases, exceptions including those for which the publisher or artist cannot be identified.

Optimized Delivery
This feature is marginally important for music files as broadband connectivity and content delivery networks (CDNs) become more and more ubiquitous, although it should be valuable for large video content for some time to come. Many DRM technologies can, with some effort, adapt to file-splitting schemes. This will be a nice-to-have feature on all kinds of online content services.

Free Content
This, of course, is not going to be possible on a copyright-respecting service. The continued presence of non-copyright-respecting networks should provide “ballast” in the market that induces copyright-respecting services to make their offerings more consumer-friendly, but (as implied above) there are many ways to do that based on features rather than price, and that trend should continue, even after any legal action takes place that puts the free file-sharing networks out of business.

Superdistribution
As mentioned above, the ready availability of e-commerce software components for such functions as payment processing, along with highly configurable DRM technology, can make Superdistribution a reality (see Rosenblatt 2003). The biggest question is whether consumers will be interested in it – i.e., interested in making the effort to resell content.

Ironically, the idea appeals most for curiosities and rarities, but if they were made available digitally, their rareness would essentially disappear. Of course, this does not take into account those who care more about collecting the physical artifacts than the actual content.

Otherwise, Superdistribution for widely-known content makes limited sense, because its only real value is as a “viral marketing” or recommendation service, in the same vein as affiliate networks like those used by Amazon.com and iTunes. If multiple participants offer the same widely-known content, then the situation devolves into one of competitive pricing, which is already the case among the many online music services that essentially offer the same products for similar prices.

New Services with P2P Influences
Even though they came into existence after the advent of Napster (the original, non-legitimate one), early copyright-respecting content services incorporated virtually none of the advantages of P2P, even when factoring out “free” vs. “pay”. Services like the US-based pressplay and the original MusicNet on RealOne featured monthly subscriptions, downloads that expired, mediocre sound quality, anemic search and browse features, no sharing, and Byzantine pricing plans seemingly borrowed from the early days of the mobile telecoms industry. Coupled with a “build it and they will come” approach to marketing, it is no wonder that critics panned these services.

Yet newer services have begun appropriating features from P2P networks. Apple’s iTunes started the trend towards offering controlled sharing. iTunes allows users to copy files onto other machines and burn MP3 versions of files onto CD limited numbers of times. US-based MusicMatch significantly raised the stakes on sharing in August 2004 by introducing a “share with your friends” feature. With this, users can send emails with playlists to as many “friends” as they like; when the friends receive the playlists, they can play the songs on them, in their entireties, up to 3 times before having to purchase them as individual downloads or subscribe to MusicMatch’s On Demand service. More recently, FNAC in France introduced its Fnacmusic download service, which raises iTunes’s 3 CD burns to 10.

Although no copyright-respecting service gives content away for free, there are a few innovative approaches to pricing in existence today. One is that of charging users a flat monthly (or annual) fee for the right to permanently download as much content as they want. One current practitioner of this model is UK-based Wippit, which is more like a modified P2P file-sharing network. Wippit maintains a list of files that are approved for sharing on the network; it enforces this not by encryption-based DRM but by a technology known as fingerprint filtering. Before a file is approved for use on the network, it is examined by a program that extracts various psycho-acoustic parameters from it in order to come with a “fingerprint” of the music in the file. The technology then searches for an instance of that fingerprint in a database of fingerprints of approved works, and if it finds a match, it lets the file go onto the network; otherwise it blocks the file.

Another alternative approach to pricing is to get users to view ads in exchange for the right to download music. Hong Kong-based Singwell International is attempting to build this type of network, which it calls Qtrax. Singwell expects to pay licensing fees to copyright holders but make revenue through its ability to sell ads that are highly targeted to users based on the kind of music they download.

A handful of new services, all US-based, are experimenting with limited forms of Superdistribution. One is Weed, a service of Seattle-based Shared Media Licensing Inc. Weed licenses independent-label music content and makes it available for purchase and eventual resale. Users can listen to Weed files up to 3 times before having to purchase them. After purchase, they can put them on websites, in emails, on CDs, or anywhere else, and pass them on to others, who can then listen to them with an option to purchase. This process can repeat arbitrarily many times. The commerce model is fixed, and it is three tiers in depth: a seller earns a 20% commission on the sale price; the user who sold it to the seller earns 10%; and the user who sold it to him earns 5%; Weed itself earns 15%, and the remaining 50% goes to the artist. Weed uses Windows Media DRM plus its own software to control this process.

Two services with multi-tier commerce models that are roughly similar to Weed are in beta at this writing. One is Bitmunk, from Virginia-based Digital Bazaar; the other is Peer Impact, from Saratoga Springs, NY, based Wurld Media. Bitmunk differs from Weed mainly in that it normally uses non-invasive watermarking instead of encryption-based DRM, which enables users to catch pirates forensically rather than preventing piracy proactively. (As is the case with some other P2P networks, Bitmunk allows users, at their own option, to put up files that are packaged with DRM.) Peer Impact combines a Weed-like commerce model with optimized content delivery (see above) a la BitTorrent. Peer Impact is unique among these services in that it has licenses, at this writing, from three of the four major recording companies.

Bottom line
While some features of P2P (such as free content) will never make it into copyright-respecting services, and other features (such as transnational usage and availability of rarities) seem highly unlikely to make it, the gaps between historically free and infringing P2P services and DRM-based copyright-respecting content services are rapidly shrinking. Over the next year or two, the boundaries of and gaps between them should become clearer through market forces and legal decisions. At the same time, the tradeoffs among new services that incorporate P2P-derived features should become more and more subtle. Content owners will need to carefully examine these services’ features as well as market forces to determine where to license their content.

Sources

About the author: Bill Rosenblatt is president of GiantSteps Media Technology Strategies, a consultancy based in New York, USA that focuses on digital content technology issues for content providers and technology vendors. He is editor of the newsletter DRM Watch (www.drmwatch.com), chair of the DRM Strategies conferences (www.drmstrategies.com), and author of Digital Rights Management: Business and Technology (John Wiley & Sons, 2001).

Status: first posted 10/12/04; included in INDICARE Monitor Vol. 1, No 6/7, 17 December 2004; licensed under Creative Commons
URL: http://www.indicare.org/tiki-read_article.php?articleId=61